BioPoint: Enhancing Human-Computer Interaction through Single-Site, Multi-Sensor Gesture Recognition.

Area of Science:

• Biomedical Engineering
• Human-Computer Interaction
• Wearable Technology

Background:

• Advancements in human-computer interaction (HCI) and machine learning aid individuals with upper limb disabilities.
• Multi-channel myoelectric systems offer intuitive user intent capture but are often bulky.
• Seamless integration of assistive technology into daily life remains a challenge.

Purpose of the Study:

• To present the first single-site EMG-IMU-PPG system for multi-gesture recognition from the wrist.
• To evaluate the system's accuracy for various finger, wrist, and hand gestures.
• To introduce "Myo Hero," an engaging online environment for testing gesture recognition systems.

Main Methods:

• Development of a compact, single-site system integrating electromyography (EMG), inertial measurement units (IMU), and photoplethysmography (PPG) sensors.
• Evaluation of the system on eight able-bodied participants performing 17 finger/wrist gestures and 8 hand gestures.
• Introduction and utilization of "Myo Hero," an interactive online platform for dynamic system evaluation.

Main Results:

• The EMG-IMU-PPG system achieved an average accuracy of 65.28%±10.45% for 17 finger and wrist gestures.
• An average accuracy of 90.98%±5.65% was obtained for 8 hand gestures.
• Online evaluation in "Myo Hero" demonstrated significantly better performance compared to random play.

Conclusions:

• The developed single-site EMG-IMU-PPG system is a viable approach for multi-gesture recognition from the wrist.
• This multimodal system shows significant potential for enhancing human-computer interaction for individuals with upper limb disabilities.
• The "Myo Hero" platform offers a novel and engaging method for evaluating gesture recognition technologies.